Browsing by Author "Kateb, Yousra"

Now showing 1 - 5 of 5

Classification of Surface Defects in Steel Sheets Using Developed NasNet-Mobile CNN and Few Samples
(IIETA, 2024) Kateb, Yousra; Khebli, Abdelmalek; Meglouli, Hocine
Rolled steel is a major product of ferrous metalworking. It is a popular metal structure construction technology. Though a big amount of the finished product may be flawed, the process of manufacturing must be improved. It is critical to correctly classify hot-rolled strip faults. As a result, in recent years, numerous machine-learning-based automated visual inspection (AVI) systems have been created. However, these approaches lack several critical components, such as insufficient RAM, which causes complexity and slowness during implementation. Long execution durations, in general, cause the process to be delayed or completed later than expected. A shortage of faulty samples is also a significant difficulty in steel defect detection, as the imbalance between the huge number of non-defective photos and the defective ones causes the algorithm to be unfair in categorization. To address these three issues, a deep CNN model is created in this study. The backbone architecture is a pre-trained NasNet-Mobile that has been fine-tuned with particular parameters to be compatible with the required data. Despite having 27 times less data than other articles' datasets, the model detects steel surface photos with six defects with 99.51% accuracy, exceeding earlier methodologies. This study is useful for surface fault classification when the sample size is small, the software is not quite as effective, or time is limited. Avoiding these issues will help the steel industry improve safety and end product quality while also saving time and money.
Classifying Surface Fault in Steel Strips Using a Customized NasNet-Mobile CNN and Small Dataset
(ESRGroups, 2024) Kateb, Yousra; Khebli, Abdelmalek; Meglouli, Hocine; Aguib, Salah; Khelifi-Touhami, Mohamed Salah
Steel metal is an important product in ferrous manufacturing, and the manufacturing process has to be improved so that hot-rolled strip flaws may be correctly identified. Machine-learning- based automated visual inspection (AVI) systems have been created, however they lack crucial components, such as inadequate RAM, resulting in complexity and sluggish implementation. Long execution times also result in delays or incompleteness. A scarcity of faulty samples further complicates steel defect diagnosis due to the disparity between non-defective and defective pictures. To overcome these difficulties, a deep CNN model is built using the pre- trained NasNet-Mobile backbone architecture. The model, which uses 26 times less data than other papers' datasets, recognizes steel surface pictures with six faults with 99.30% accuracy, outperforming previous methods. This study is beneficial for surface fault classification when the sample size is small, the software is less effective, or time is limited. Avoiding these issues will improve safety and end product quality in the steel industry, saving time and money
Coronavirus Diagnosis Based on Chest X-Ray Images and Pre-Trained DenseNet-121
(IIETA, 2023) Kateb, Yousra; Meglouli, Hocine; Khebli, Abdelmalek
A serious global problem called COVID-19 has killed a great number of people and rendered many projects useless. The obtained individual's identification at the appropriate time is one of the crucial methods to reduce losses. By detecting and recognizing contaminated individuals in the early stages, artificial intelligence can help many associations in these situations. In this study, we offer a fully automated method to identify COVID-19 from a patient's chest X-ray images without the need for a clinical expert's assistance. The proposed approach was evaluated on the public COVID-19 X-ray dataset that achieves high performance and reduces computational complexity. This dataset contains 400 photos, 100 images of individuals who were infected with Covid-19, 100 images of individuals with no COVID-19, 100 images of a viral pneumonia and a 100 more images that we reserve them for testing part. So we have an overall 300 images for training and 100 for testing. The obtained results were so satisfying, an F1 score of 0.98, a Recall of 0.98, and an Accuracy of 0.98. The classification method deep learning-based DenseNet-121, transfer learning, as well as data augmentation techniques were implemented to improve the model more accurately. Our proposed approach outperforms several CNNs and all recent works on COVID‑19 images. Even though there are not enough training photos comparing to other extra-large datasets.
Intelligent monitoring of an industrial system using image classification
(Universite M'Hamed Bougara Boumerdès : Faculté des Hydrocarbures et de la Chimie, 2025) Kateb, Yousra; Meglouli, Hocine(Directeur de thèse)
L'importance de l'automatisation pour améliorer l'efficacité et la rentabilité, tout en réduisant au minimum la participation humaine, a entraîné une augmentation considérable des industries automatisées dans divers secteurs à l'échelle mondiale. Cependant, ces industries doivent surmonter un certain nombre d'obstacles pour trouver des défauts dans le produit fini. Dans ce scénario particulier, la mise en oeuvre de systèmes de surveillance intelligents semble contribuer grandement à l'inspection automatisée des défauts de la qualité des produits tout en réduisant la nécessité d'une intervention humaine excessive. Cependant, ces systèmes manquent de plusieurs composants essentiels, ce qui entraîne des retards et des inefficacités de la production. Le domaine fait face à des défis importants, notamment un temps d'exécution limité, des coûts élevés et l'utilisation de processeurs qui manquent de puissance suffisante à des fins de surveillance. Cette thèse utilise un réseau neuronal convolutionel récemment développé, NasNet-Mobile, dans une tentative de surmonter ces limitations. Notre étude a impliqué l'utilisation d'un réseau neuronal profond cusomisé qui a fait l'objet d'une formation approfondie pour classer avec précision six défauts distincts trouvés sur les surfaces d'acier. L'algorithme a été amélioré en utilisant la fonction d'activation de l'ELU au lieu de ReLU pour résoudre le problème des neurones en train de mourir. Les optimisateurs ont été changés entre ADAM et ADAMAX pour trouver le meilleur. De nombreuses autres couches ont été ajoutées pour améliorer l'apprentissage du modèle, comme Dropout et le Global Average Pooling. Les résultats ont été très prometteurs, car l'approche proposée a réussi à classer six types différents d'images défectueuses dans le domaine de l'acier industriel. Même avec beaucoup moins de données que d'autres ensembles de données de recherche, le modèle a obtenu une précision impressionnante de 99,51 % avant le réglage fin et de 100 % après le reglage fin, dépassant les méthodes précédentes d'évaluation des défauts de pointe. L'importance de cette recherche réside dans son potentiel d'améliorer la surveillance intelligente de l'industrie dans des scénarios difficiles, tels que des échantillons d'images défectueux limités, des logiciels moins puissants ou des situations sensibles au temps. En répondant à ces défis, l'industrie de l'acier peut réduire les coûts et le temps consacré à l'ensemble du processus tout en améliorant la sécurité des travailleurs et la qualité des produits
Steel surface defect detection using convolutional neural network
(2020) Kateb, Yousra; Meglouli, Hocine; Khebli, Abdelmalek
Steel is the most important engineering and construction material in the world. It is used in all aspects of our lives. But as every metal is can be defected and then will not be useful by the consumer Steel surface inspection has seen an important attention in relation with industrial quality of products. In addition, it has been studied in different methods based on image classification in the most of time, but these can detect only such kind of defects in very limited conditions such as illumination, obvious contours, contrast and noise...etc. In this paper, we aim to try a new method to detect steel defects this last depend on artificial intelligence and artificial neural networks. We will discuss the automatic detection of steel surface defects using the convolutional neural network, which can classify the images in their specific classes. The steel we are going to use will be well-classified weather the conditions of imaging are not the same, and this is the advantage of the convolutional neural network in our work. The accuracy and the robustness of the results are so satisfying